# Found a BUG and basic docs code fails to run on kaggle tpu

**URL:** <https://discuss.huggingface.co/t/found-a-bug-and-basic-docs-code-fails-to-run-on-kaggle-tpu/55153>\
**Category:** 🤗Accelerate\
**Created:** [September 15, 2023, 4:38pm UTC](https://discuss.huggingface.co/t/found-a-bug-and-basic-docs-code-fails-to-run-on-kaggle-tpu/55153 "2023-09-15T16:38:02Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![manoj333](https://avatars.discourse-cdn.com/v4/letter/m/9dc877/32.png) [@manoj333](https://discuss.huggingface.co/u/manoj333)\
**Post date:** [September 15, 2023, 4:38pm UTC](https://discuss.huggingface.co/t/found-a-bug-and-basic-docs-code-fails-to-run-on-kaggle-tpu/55153/1 "2023-09-15T16:38:02Z")

</div>

So just to try TPU, i’ve used the code by @muellerzr [here](https://huggingface.co/blog/pytorch-ddp-accelerate-transformers) on kaggle tpu vm, it gives me  
the error `BrokenProcessPool: A process in the process pool was terminated abruptly while the future was running or pending.` [link to traceback and code](https://filebin.net/7ocy5d8s6ndgnans)  
Any help on this is appreciated.

**BUG**  
Also accelerate uses os.environ variable TPU\_NAME to check if its tpu or not and it won’t work here,and this leads to trying to run 8 gpu’s in a TPU environment 🥲